MORTALITY RISK BY BODY-WEIGHT AND WEIGHT CHANGE IN PEOPLE WITH NIDDM - THE WHO MULTINATIONAL STUDY OF VASCULAR-DISEASE IN DIABETES

Citation
N. Chaturvedi et al., MORTALITY RISK BY BODY-WEIGHT AND WEIGHT CHANGE IN PEOPLE WITH NIDDM - THE WHO MULTINATIONAL STUDY OF VASCULAR-DISEASE IN DIABETES, Diabetes care, 18(6), 1995, pp. 766-774
Citations number
44
Categorie Soggetti
Endocrynology & Metabolism","Medicine, General & Internal
Journal title
ISSN journal
01495992
Volume
18
Issue
6
Year of publication
1995
Pages
766 - 774
Database
ISI
SICI code
0149-5992(1995)18:6<766:MRBBAW>2.0.ZU;2-W
Abstract
OBJECTIVE - Care guidelines for people with non-insulin-dependent diab etes mellitus (NIDDM) emphasize the importance of weight loss in reduc ing mortality risk. However, existing evidence regarding the relations hip between weight and mortality and the effects of weight change is c onflicting. We examined these relationships in the World Health Organi zation Multinational Study of Vascular Disease in Diabetes. RESEARCH D ESIGN AND METHODS - This was a cohort study of 1,416 men and 1,544 wom en. Baseline examinations were performed in 1975 through 1977, a morbi dity follow-up was performed in 1983, and a mortality follow-up contin ued until 1988. Data were analyzed according to geographical groups: E uropeans, East Asians, and Native Americans. The relationship between weight change and mortality was analyzed for Europeans only. RESULTS - Generally, body mass index (BMI) was positively associated with age, blood pressure, and cholesterol but was negatively associated with dur ation of diabetes, prevalence of retinopathy, and use of insulin. Ther e was no clear relationship between BMI and mortality across the geogr aphical groups. In Europeans, weight loss in the leanest subjects at b aseline (BMI < 26 kg/m(2)) was associated with a threefold increase in mortality risk compared with those who had maintained a steady weight (relative risk [RR] 3.05, 95% confidence interval [CI] 1.26-7.36). On ly in the most obese group was weight loss associated with a reduction in mortality risk (BMI > 29 kg/m(2), RR 0.84, 95% CI 0.40-1.74). CONC LUSIONS - The positive association of BMI with age, blood pressure, an d cholesterol and the negative association with duration of diabetes, retinopathy, and use of insulin may explain why there is no strong rel ationship between BMI and mortality in NIDDM. Weight loss, particularl y in the relatively lean diabetic person, may be associated with an in creased mortality risk.